Inspiration
We were inspired to create this project as we have witnessed people handling kitchen equipment very unsafely. While they may have cut the goods, there were major risks to their own health, as knife safety is very important. We looked into why we believe this is, and we have come to the conclusion that part of this root issue stems from cookbooks, in which one is told to just prepare the produce, without any explanation on the how.
What it does
CutSmart is a mobile app allows for the user to snap a photo of a piece of produce in front of them, then they will be told some safety tips on it. Then, there is a method in order to let them watch a video on how to perform proper cutting techniques. Furthermore, we have implemented a history page and login, which allow for the user to keep track of all the items they have scanned in the past.
When scanning an item, an ai program will try to decipher what it is, then gives the user the chance to retake a photo if it is incorrect. If it is correct, they will be directed to a page wherein there will be some nutritional values, and safety tips
How we built it
We coded in android studio using flutter and dart for our main project, however, we connected our MongoDB database using python. Dart innately allows for the implementation of Google gemini. For our prompt in gemini, we let it create a json file based on it's interpretation of the picture, then took the key details to give to the user. We coded this starting in scenes with transitions, then we built upon each necessary scene. We then simplified this using classes for repetitive parts, primarily the home bar which will allow for navigation
Challenges we ran into
Accomplishments that we're proud of
What we learned
We learned how to do mobile app development, as it was all of our first times. We learned a tremendous amount about git as well, as we ran into branching and merging issues. This led us to deepening our fundamental knowledge, while also proving useful to the project. Furthermore, we learned how to use AI apis, alongside connecting to databases. While it seems quite simple at first, there are many nuances that can set one back by a few hours. Paired with the time constraint of this project, we overall found a way to balance efficient coding without needing to backtrack every few moments
What's next for CutSmart
We want to fully deploy this mobile app, while cleaning up some of the minor consistencies. We believe our database is still underutilized, and that there is much more we can do with it. This includes the ability to come back to previous explanations without having to rescan the produce. Furthermore, this code has many efficiency issues, and thus runs at a rate much slower than possible. We want to fix these issues, as we believe consumer experience relates directly to not only functionality, but the convince.
Log in or sign up for Devpost to join the conversation.